Estimating Differentiated Goods Oligopoly Models with Market Share and Cost Data

In this paper, we propose a methodology of jointly estimating the demand and cost functions of the differentiated goods oligopoly models when there are both data on demand and cost of firms. An important characteristics of the method is that it deals both with endogeneity of prices from the unobserved quality term, and the endogeneity of output in the cost function from the unobserved cost shock. Our method will do so without using any instruments. We establish identification and consistency of our estimator. In Monte-Carlo exercises, demonstrate that it works well in situations where the instruments are correlated with the error term, and thus the standard IV approach results in bias.